基于投影滤波的突发事件危险源位置估计问题研究

针对确定突发事件中危险源的位置估计问题,提出一种基于hermit基函数的投影滤波方法.通过将其概率解投影到指定状态空间中,通过投影空间中的艾尔米特基函数构造出近似解,即状态的先验概率密度函数,最后由贝叶斯估计得到状态的后验概率密度函数,给出算法的收敛性和计算复杂度分析.仿真实验证明投影滤波方法有效减少滤波估计中所需的计算量,并提高了估计精度.

Hazard position estimation of sudden accidents based on projection filter

In this paper, a method applying projection filter algorithm based on hermit primary function has been proposed, which is focusing on dealing with the hazard position estimation of sudden accidents problem. By constructing an approximate solution with the hermit primary function in the projection space, the new algorithm could approximate the nonlinear filter result by solving for the prior probability density, then a posterior probability density function is gained through Bayesian formula, finally, the performance of convergence and computational complexity are considered and discussed. Simulating experiments results show that the new projection filter method can improve the estimating accuracy and reduce calculation.